Robert Cookson, Digital Media Correspondent reported in the Financial Times (Sept. 22, 2015) that, "Google has been charging marketers for advertisements it places on YouTube even when the video platform’s fraud-detection systems identify that a “viewer” is a robot rather than a human being." http://www.ft.com/cms/s/0/53ac3fd0-604e-11e5-a28b-50226830d644.html
Cookson's reportage is based on a research paper (http://arxiv.org/pdf/1507.08874v1.pdf) published July 31, 2015 by a group of seven European researchers. Their key finding when looking at several different video portals paraphrased by Cookson, "Google’s core advertising engine charged the researchers for the bot visits even though YouTube was clearly able to identify them as fake."
The research concludes, "In summary, we conclude that YouTube uses a seemingly permissive detection mechanism to discount fake monetized views. This exposes advertisers to... burden the risk of fraud. [While] Conversely, the public view counter is much more discriminative, demonstrating that YouTube has effective means to identify fake views."
The channel/network that receives the fake views is also penalized, "YouTube severely penalizes the uploader of the video by suspending her Adsense account, preventing the uploader from monetizing any of the videos associated to the suspended account" even when some of them were legitimate.
Researchers:
Miriam Marciel NEC Labs Europe [email protected]
Ruben Cuevas Universidad Carlos III de Madrid [email protected]
Albert Banchs IMDEA Networks Institute Universidad Carlos III de Madrid [email protected]
Roberto Gonzalez NEC Labs Europe [email protected]
Stefano Traverso Politecnico di Torino [email protected]
Mohamed Ahmed NEC Labs Europe [email protected]
Arturo Azcorra IMDEA Networks Institute Universidad Carlos III de Madrid [email protected]